Re: [PATCH 3/4] ptrace: remove PT_DTRACE from m68k, m68knommu

[Geert Uytterhoeven <geert@linux-m68k.org>]

On Fri, Apr 24, 2009 at 12:06, Andreas Schwab <schwab@linux-m68k.org> wrote:
> Oleg Nesterov <oleg@redhat.com> writes:
>
>> m68k sets PT_DTRACE in trap_c() but never uses it.
>>
>> Signed-off-by: Oleg Nesterov <oleg@redhat.com>
>>
>>  arch/m68k/kernel/traps.c            |    1 -
>>  arch/m68knommu/kernel/asm-offsets.c |    1 -
>>  arch/m68knommu/kernel/traps.c       |    1 -
>>  3 files changed, 3 deletions(-)
>>
>> --- PTRACE/arch/m68k/kernel/traps.c~DT_3_SET  2009-04-06 00:03:35.000000000 +0200
>> +++ PTRACE/arch/m68k/kernel/traps.c   2009-04-22 21:52:23.000000000 +0200
>> @@ -1057,7 +1057,6 @@ asmlinkage void trap_c(struct frame *fp)
>>       if (fp->ptregs.sr & PS_S) {
>>               if ((fp->ptregs.vector >> 2) == VEC_TRACE) {
>>                       /* traced a trapping instruction */
>> -                     current->ptrace |= PT_DTRACE;
>
> This is supposed to be setting the TIF_DELAYED_TRACE thread flag.

This code already looks different in the m68k tree, due to your fix to
`Allow all kernel traps to be handled via exception fixups' for
ARAnyM, and some related cleanups by you and Roman.

But more interestingly, Roman's last cleanup said there's no need to
set TIF_DELAYED_TRACE?

Gr{oetje,eeting}s,

						Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org

In personal conversations with technical people, I call myself a hacker. But
when I'm talking to journalists I just say "programmer" or something like that.
							    -- Linus Torvalds